Ionosphere TEC Model using General Regression Neural Network in Eastern Indonesia

Total Electron Content (TEC) data can be obtained from Global Navigation Satellite Systems (GNSS) data. The GNSS receiver can record the ionosphere conditions when radiating in the ionosphere layer. Because the dispersive ionosphere layer, the radio waves emitted by GNSS will be affected differently depending on the frequency used. The lower the frequency, the greater the effect of ionospheric bias. So the GNSS signal can record the existing ionosphere conditions. Currently there are many methods used for TEC modeling of GNSS data, both global and regional models. One of them is artificial neural network method. Artificial neural network modeling can be developed into a general regression neural network (GRNN). TEC model can be used to determine the existence of an ionospheric storm caused by a geomagnetic storm if the TEC input (of each receiver) contains a geomagnetic signal. This study was TEC modeling use GRNN method. The data used are GNSS 24 continuous stations located in eastern Indonesia. GRNN TEC modeling is applied during geomagnetic storm, before and after. TEC modeling is done to see the effect of geomagnetic storm on the ionosphere occurring in eastern Indonesia. The GRNN TEC modeling with optimum model constants ranging from 0.1 to 100 indicates an ionosphere storm occurring in eastern Indonesia in October 2016